From Specification to Reality
A standard is only valuable if providers actually implement it. As of 2024, FOCUS has achieved meaningful industry adoption:
- AWS supports FOCUS-compliant exports through AWS Data Exports (the successor to Cost and Usage Reports). You can configure an export in the AWS Billing console to output FOCUS 1.0 format directly to S3.
- Microsoft Azure supports FOCUS exports through Azure Cost Management. You can configure a scheduled export to a storage account in FOCUS format.
- Google Cloud has committed to FOCUS support through BigQuery billing exports.
- Oracle Cloud and several SaaS vendors (including some Kubernetes cost tools) have also committed to or implemented FOCUS exports.
Enabling FOCUS Exports on AWS (Reference)
In the AWS console, navigate to Billing → Data Exports → Create Export. Select "FOCUS 1.0" as the export format. The resulting S3 files will use FOCUS column names instead of the legacy CUR column names.
The key difference from legacy CUR:
- Legacy CUR: lineItem/UnblendedCost, lineItem/UsageStartDate, product/region
- FOCUS export: BilledCost, ChargePeriodStart, RegionId
What FOCUS Does Not Solve
FOCUS standardizes the schema of billing data, not the semantics of every service. Some nuances remain provider-specific:
- Pricing models: AWS Spot Instances, Azure Spot VMs, and GCP Preemptible VMs are all represented as
Computeusage, but their pricing volatility and interruption behavior differ. FOCUS captures the cost but not the risk profile. - Service granularity:
ServiceNamemaps to high-level service names. Distinguishing between an EC2t3.microand ac5.4xlargestill requires looking atResourceTypeor SKU-level columns. - Commitment optimization: FOCUS tells you what you spent; it does not tell you whether your Savings Plan coverage is optimal. That analysis still requires provider-specific tooling or FinOps platforms.
FOCUS is a foundation, not a complete solution. It eliminates the translation problem so that tools and analysts can focus on the analysis problem.
If asked "What are the limitations of FOCUS?", avoid saying it solves everything. A strong answer acknowledges that FOCUS standardizes schema but not semantics — you still need provider-specific knowledge to interpret why costs are what they are, optimize commitment coverage, or understand service-level pricing models. This shows you understand the difference between data normalization and data interpretation.